Merge pull request #228 from k3s-io/default_k3s_script

Fix HA, simplify provisioning, add Vagrant test cluster

.ansible-lint

@@ -0,0 +1,5 @@
+---
+warn_list:
+ - var-naming[no-role-prefix]
+ - yaml[comments-indentation]
+ - yaml[line-length]

README.md

@@ -22,7 +22,7 @@ on processor architecture:
 ## System requirements
 
 Deployment environment must have Ansible 2.4.0+
-Master and nodes must have passwordless SSH access
+Server and agent nodes must have passwordless SSH access
 
 ## Usage
 
@@ -48,7 +48,7 @@ k3s_cluster:
 If needed, you can also edit `vars` section at the bottom to match your environment.
 
 If multiple hosts are in the server group the playbook will automatically setup k3s in HA mode with embedded etcd.
-An odd number of server nodes is recommended (3,5,7). Read the offical documentation below for more information and options.
+An odd number of server nodes is required (3,5,7). Read the offical documentation below for more information and options.
 https://rancher.com/docs/k3s/latest/en/installation/ha-embedded/
 Using a loadbalancer or VIP as the API endpoint is preferred but not covered here.
 
@@ -61,8 +61,19 @@ ansible-playbook playbook/site.yml -i inventory.yml
 
 ## Kubeconfig
 
-To confirm access to your **Kubernetes** cluster use the following:
+After successful bringup, the kubeconfig of the cluster is copied to the control-node and set as default (`~/.kube/config`).
+Assuming you have [kubectl](https://kubernetes.io/docs/tasks/tools/#kubectl) installed, you to confirm access to your **Kubernetes** cluster use the following:
 
 ```bash
 kubectl get nodes
 ```
+
+## Local Testing
+
+A Vagrantfile is provided that provision a 5 nodes cluster using LibVirt or Virtualbox and Vagrant. To use it:
+
+```bash
+vagrant up
+```
+
+By default, each node is given 2 cores and 2GB of RAM and runs Ubuntu 20.04. You can customize these settings by editing the `Vagrantfile`.

Vagrantfile

@@ -0,0 +1,56 @@
+# ENV['VAGRANT_NO_PARALLEL'] = 'no'
+NODE_ROLES = ["server-0", "server-1", "server-2", "agent-0", "agent-1"]
+NODE_BOXES = ['generic/ubuntu2004', 'generic/ubuntu2004', 'generic/ubuntu2004', 'generic/ubuntu2004', 'generic/ubuntu2004']
+NODE_CPUS = 2
+NODE_MEMORY = 2048
+# Virtualbox >= 6.1.28 require `/etc/vbox/network.conf` for expanded private networks 
+NETWORK_PREFIX = "10.10.10"
+
+def provision(vm, role, node_num)
+  vm.box = NODE_BOXES[node_num]
+  vm.hostname = role
+  # We use a private network because the default IPs are dynamicly assigned 
+  # during provisioning. This makes it impossible to know the server-0 IP when 
+  # provisioning subsequent servers and agents. A private network allows us to
+  # assign static IPs to each node, and thus provide a known IP for the API endpoint.
+  node_ip = "#{NETWORK_PREFIX}.#{100+node_num}"
+  # An expanded netmask is required to allow VM<-->VM communication, virtualbox defaults to /32
+  vm.network "private_network", ip: node_ip, netmask: "255.255.255.0"
+
+  vm.provision "ansible", run: 'once' do |ansible|
+    ansible.compatibility_mode = "2.0"
+    ansible.playbook = "playbook/site.yml"
+    ansible.groups = {
+      "server" => NODE_ROLES.grep(/^server/),
+      "agent" => NODE_ROLES.grep(/^agent/),
+      "k3s_cluster:children" => ["server", "agent"],
+    }
+    ansible.extra_vars = {
+      k3s_version: "v1.26.5+k3s1",
+      api_endpoint: "#{NETWORK_PREFIX}.100",
+      token: "myvagrant",
+      # Required to use the private network configured above
+      extra_server_args: "--node-external-ip #{node_ip} --flannel-iface eth1", 
+      extra_agent_args: "--node-external-ip #{node_ip} --flannel-iface eth1",
+    }
+  end
+end
+
+Vagrant.configure("2") do |config|
+  # Default provider is libvirt, virtualbox is only provided as a backup
+  config.vm.provider "libvirt" do |v|
+    v.cpus = NODE_CPUS
+    v.memory = NODE_MEMORY
+  end
+  config.vm.provider "virtualbox" do |v|
+    v.cpus = NODE_CPUS
+    v.memory = NODE_MEMORY
+  end
+
+  NODE_ROLES.each_with_index do |name, i|
+    config.vm.define name do |node|
+      provision(node.vm, name, i)
+    end
+  end
+
+end

inventory-sample.yml

@@ -9,14 +9,17 @@ k3s_cluster:
         192.16.35.12
         192.16.35.13
 
+  # Required Vars
   vars:
     ansible_port: 22
     ansible_user: debian
     k3s_version: v1.25.5+k3s2
-    k3s_server_location: /var/lib/rancher/k3s
-    systemd_dir: /etc/systemd/system
+    token: "mytoken"  # Use ansible vault if you want to keep it secret
     api_endpoint: "{{ hostvars[groups['server'][0]]['ansible_host'] | default(groups['server'][0]) }}"
-    api_port: 6443
     extra_server_args: ""
-    extra_server_init_args: ""
     extra_agent_args: ""
+
+  # Optional vars
+    # api_port: 6443
+    # k3s_server_location: /var/lib/rancher/k3s
+    # systemd_dir: /etc/systemd/system

roles/download/tasks/main.yml

@@ -1,38 +1,17 @@
 ---
-- name: Download k3s binary x64
+- name: Download k3s install script
   ansible.builtin.get_url:
-    url: https://github.com/k3s-io/k3s/releases/download/{{ k3s_version }}/k3s
-    checksum: sha256:https://github.com/k3s-io/k3s/releases/download/{{ k3s_version }}/sha256sum-amd64.txt
+    url: https://get.k3s.io/
     timeout: 120
-    dest: /usr/local/bin/k3s
+    dest: /usr/local/bin/k3s-install.sh
     owner: root
     group: root
     mode: 0755
-  when: ansible_facts.architecture == "x86_64"
 
-- name: Download k3s binary arm64
-  ansible.builtin.get_url:
-    url: https://github.com/k3s-io/k3s/releases/download/{{ k3s_version }}/k3s-arm64
-    checksum: sha256:https://github.com/k3s-io/k3s/releases/download/{{ k3s_version }}/sha256sum-arm64.txt
-    timeout: 120
-    dest: /usr/local/bin/k3s
-    owner: root
-    group: root
-    mode: 0755
-  when:
-    - ( ansible_facts.architecture is search("arm") and
-        ansible_facts.userspace_bits == "64" ) or
-      ansible_facts.architecture is search("aarch64")
-
-- name: Download k3s binary armhf
-  ansible.builtin.get_url:
-    url: https://github.com/k3s-io/k3s/releases/download/{{ k3s_version }}/k3s-armhf
-    checksum: sha256:https://github.com/k3s-io/k3s/releases/download/{{ k3s_version }}/sha256sum-arm.txt
-    timeout: 120
-    dest: /usr/local/bin/k3s
-    owner: root
-    group: root
-    mode: 0755
-  when:
-    - ansible_facts.architecture is search("arm")
-    - ansible_facts.userspace_bits == "32"
+- name: Download k3s binary
+  ansible.builtin.command:
+    cmd: /usr/local/bin/k3s-install.sh
+  environment:
+    INSTALL_K3S_SKIP_START: "true"
+    INSTALL_K3S_VERSION: "{{ k3s_version }}"
+  changed_when: true

roles/k3s/agent/defaults/main.yml

@@ -0,0 +1,4 @@
+---
+k3s_server_location: "/var/lib/rancher/k3s"
+systemd_dir: "/etc/systemd/system"
+api_port: 6443

roles/k3s/agent/tasks/main.yml

@@ -1,4 +1,5 @@
 ---
+
 - name: Copy K3s service file
   ansible.builtin.template:
     src: "k3s-agent.service.j2"
@@ -11,5 +12,5 @@
   ansible.builtin.systemd:
     name: k3s-agent
     daemon_reload: true
-    state: restarted
+    state: started
     enabled: true

roles/k3s/agent/templates/k3s-agent.service.j2

@@ -1,13 +1,17 @@
 [Unit]
 Description=Lightweight Kubernetes
 Documentation=https://k3s.io
+Wants=network-online.target
 After=network-online.target
 
+[Install]
+WantedBy=multi-user.target
+
 [Service]
 Type=notify
-ExecStartPre=-/sbin/modprobe br_netfilter
-ExecStartPre=-/sbin/modprobe overlay
-ExecStart=/usr/local/bin/k3s agent --data-dir {{ k3s_server_location }} --server https://{{ api_endpoint }}:{{ api_port }} --token {{ hostvars[groups['server'][0]]['token'] }} {{ extra_agent_args }}
+EnvironmentFile=-/etc/default/%N
+EnvironmentFile=-/etc/sysconfig/%N
+EnvironmentFile=-/etc/systemd/system/k3s.service.env
 KillMode=process
 Delegate=yes
 # Having non-zero Limit*s causes performance problems due to accounting overhead
@@ -19,6 +23,7 @@ TasksMax=infinity
 TimeoutStartSec=0
 Restart=always
 RestartSec=5s
-
-[Install]
-WantedBy=multi-user.target
+ExecStartPre=/bin/sh -xc '! /usr/bin/systemctl is-enabled --quiet nm-cloud-setup.service'
+ExecStartPre=-/sbin/modprobe br_netfilter
+ExecStartPre=-/sbin/modprobe overlay
+ExecStart=/usr/local/bin/k3s agent --data-dir {{ k3s_server_location }} --server https://{{ api_endpoint }}:{{ api_port }} --token {{ token }} {{ extra_agent_args }}

roles/k3s/server/defaults/main.yml

@@ -0,0 +1,4 @@
+---
+k3s_server_location: "/var/lib/rancher/k3s"
+systemd_dir: "/etc/systemd/system"
+api_port: 6443

roles/k3s/server/tasks/main.yml

@@ -2,50 +2,30 @@
 - name: Init first server node
   when: ansible_hostname == groups['server'][0]
   block:
-    - name: Start temporary service for HA cluster
-      ansible.builtin.command:
-        cmd: >
-          systemd-run -p RestartSec=2 -p Restart=on-failure --unit=k3s-init k3s server
-          --cluster-init --tls-san {{ api_endpoint }} --data-dir {{ k3s_server_location }} {{ extra_server_args}}
-        # noqa: jinja[spacing]
-        creates: "{{ k3s_server_location }}/server/node-token"
-      when: groups['server'] | length > 1
-
-    - name: Start temporary service for single server cluster
-      ansible.builtin.command:
-        cmd: >
-          systemd-run -p RestartSec=2 -p Restart=on-failure --unit=k3s-init k3s server
-           --tls-san {{ api_endpoint }} --data-dir {{ k3s_server_location }} {{ extra_server_args }}
-        creates: "{{ k3s_server_location }}/server/node-token"
+    - name: Copy K3s service file [Single]
       when: groups['server'] | length == 1
-
-    - name: Wait for node-token
-      ansible.builtin.wait_for:
-        path: "{{ k3s_server_location }}/server/node-token"
-
-    - name: Register node-token file access mode
-      ansible.builtin.stat:
-        path: "{{ k3s_server_location }}/server/node-token"
-      register: p
-
-    - name: Change file access node-token
-      ansible.builtin.file:
-        path: "{{ k3s_server_location }}/server/node-token"
-        mode: "g+rx,o+rx"
-
-    - name: Read node-token from server
-      ansible.builtin.slurp:
-        path: "{{ k3s_server_location }}/server/node-token"
-      register: node_token
-
-    - name: Store server node-token
-      ansible.builtin.set_fact:
-        token: "{{ node_token.content | b64decode | regex_replace('\n', '') }}"
-
-    - name: Restore node-token file access
-      ansible.builtin.file:
-        path: "{{ k3s_server_location }}/server/node-token"
-        mode: "{{ p.stat.mode }}"
+      ansible.builtin.template:
+        src: "k3s-single.service.j2"
+        dest: "{{ systemd_dir }}/k3s.service"
+        owner: root
+        group: root
+        mode: 0644
+
+    - name: Copy K3s service file [HA]
+      when: groups['server'] | length > 1
+      ansible.builtin.template:
+        src: "k3s-cluster-init.service.j2"
+        dest: "{{ systemd_dir }}/k3s.service"
+        owner: root
+        group: root
+        mode: 0644
+
+    - name: Enable and check K3s service
+      ansible.builtin.systemd:
+        name: k3s
+        daemon_reload: true
+        state: started
+        enabled: true
 
     - name: Create directory .kube
       ansible.builtin.file:
@@ -54,6 +34,10 @@
         owner: "{{ ansible_user }}"
         mode: "u=rwx,g=rx,o="
 
+    - name: Pause to allow server startup
+      ansible.builtin.pause:
+        seconds: 10
+
     - name: Copy config file to user home directory
       ansible.builtin.copy:
         src: /etc/rancher/k3s/k3s.yaml
@@ -76,17 +60,28 @@
         flat: true
 
 - name: Start other server if any and verify status
+  when:
+    - (groups['server'] | length) > 1
+    - ansible_hostname != groups['server'][0]
   block:
-    - name: Init additonal server nodes
-      ansible.builtin.command:
-        cmd: >
-          systemd-run -p RestartSec=2 -p Restart=on-failure --unit=k3s-init k3s server
-          --token "{{ hostvars[groups['server'][0]]['token'] }}" --server https://{{ api_endpoint }}:{{ api_port }}
-          --tls-san {{ api_endpoint }} --data-dir {{ k3s_server_location }} {{ extra_server_args }}
-        creates: "{{ k3s_server_location }}/server/node-token"
-      when: ansible_hostname != groups['server'][0]
+    - name: Copy K3s service file [HA]
+      when: groups['server'] | length > 1
+      ansible.builtin.template:
+        src: "k3s-ha.service.j2"
+        dest: "{{ systemd_dir }}/k3s.service"
+        owner: root
+        group: root
+        mode: 0644
+
+    - name: Enable and check K3s service
+      ansible.builtin.systemd:
+        name: k3s
+        daemon_reload: true
+        state: started
+        enabled: true
 
     - name: Verify that all server nodes joined
+      when: (groups['server'] | length) > 1
       ansible.builtin.command:
         cmd: >
           k3s kubectl get nodes -l "node-role.kubernetes.io/control-plane=true" -o=jsonpath="{.items[*].metadata.name}"
@@ -95,28 +90,6 @@
       retries: 20
       delay: 10
       changed_when: false
-  always:
-    - name: Kill the temporary init service
-      ansible.builtin.systemd:
-        name: k3s-init
-        state: stopped
-      failed_when: false
-
-- name: Copy K3s service file
-  ansible.builtin.template:
-    src: "k3s-server.service.j2"
-    dest: "{{ systemd_dir }}/k3s-server.service"
-    owner: root
-    group: root
-    mode: 0644
-  register: k3s_service
-
-- name: Enable and check K3s service
-  ansible.builtin.systemd:
-    name: k3s-server
-    daemon_reload: true
-    state: restarted
-    enabled: true
 
 - name: Create symlinks
   ansible.builtin.file: